Comparison of running power algorithms across different running speeds in 1,550 recreationally active adults.

Authors

Gustavo Leporace, Thiago F Shida, Nicholas Metsavaht, Eliane C Guadagnin, Leonardo Metsavaht

Published in

Journal of biomechanics. Volume 205. Pages 113470. Jul 14, 2026. Epub Jul 14, 2026.

Abstract

Running mechanical power can be estimated using formulations that quantify different components of mechanical work, including external center-of-mass dynamics and internal segmental work. Several algorithms have been proposed to estimate running power from kinematic data or theoretical analytical formulations, yet most have been evaluated in small samples of healthy runners, limiting generalizability. This study aimed to compare four mechanical power algorithms (Minetti; Theoretical Power Work 1 (TPw1); Theoretical Power Work 2 (TPw2); and Jenny & Jenny (JJ)) in a large and heterogeneous cohort of recreationally active adults, and to determine whether algorithm-velocity relationships were consistent in men and women. Three-dimensional kinematic data from 1,550 participants (39.9 ± 10.6 years; 725 women, 825 men) were collected during treadmill running using a motion capture system. Running power was computed for each model in both absolute (W) and relative (W·kg⁻1) terms. Correlations with treadmill velocity were assessed using Spearman's rank test, and between-model differences were evaluated through one-way ANOVA and effect sizes (Cohen's d). Normalized power values exhibited near-perfect correlations with velocity for Minetti, TPw1, and TPw2 (r = 0.964-0.968, r² = 0.929-0.938), whereas JJ showed slightly lower but still high correlations (r = 0.914, r² = 0.835). All between-model differences were significant (p < 0.001), with large effects (d > 0.8) observed mainly between Minetti and TPw2. These findings demonstrated that Minetti, TPw1, and TPw2 yield consistent estimates of running power, with mass-normalized outputs showing robust power-velocity relationships across a heterogeneous sample of recreationally active adults from both sexes, while model selection should consider the analytical goal and data availability.

PMID:
42456207
Bibliographic data and abstract were imported from PubMed on 16 Jul 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.